Group Title: relationship of violations of receiver expectations and resistance to persuasion /
Title: The relationship of violations of receiver expectations and resistance to persuasion
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097474/00001
 Material Information
Title: The relationship of violations of receiver expectations and resistance to persuasion
Physical Description: iv, 77 leaves : ; 28 cm.
Language: English
Creator: Miller, Michael David, 1952-
Publication Date: 1978
Copyright Date: 1978
 Subjects
Subject: Persuasion (Psychology)   ( lcsh )
Speech thesis Ph. D   ( lcsh )
Dissertations, Academic -- Speech -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 72-75.
Additional Physical Form: Also available on World Wide Web
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Michael D. Miller.
 Record Information
Bibliographic ID: UF00097474
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000074585
oclc - 04695805
notis - AAH9859

Downloads

This item has the following downloads:

PDF ( 2 MBs ) ( PDF )


Full Text
















THE RELATIONSHIP OF VIOLATIONS OF RECEIVER EXPECTATIONS
AND RESISTANCE TO PERSUASION










By

Michael D. Miller


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY






UNIVERSITY OF FLORIDA


1978














ACKNOWLEDGEMENTS


I would like to acknowledge two very important people: Miichael

Burgoon and Debra Miller. Through Michael my way of thinking has

changed; through Debra, my way of feeling. Because of these two

people I no longer view the world in the same way.














TABLE OF CONTENTS


ACKNOWLEDGEMENTS . . . . . . . . . . .

ABSTRACT . . . . . . . . . . . . .

CHAPTER

I RATIONALE . . . . . . . . . .

Threat and Counterarguing . . . . . .
Threat and Violations of Expectations . . .
A New Model of Resistance to Persuasion .. ...
Derivation of the Hypotheses . . . . .

II METHODS . . . . . . . . . . .

Design of the Experiment . . . . . .
Message Construction . . . . . . .
Experimental Procedures . . . . . . .

III RESULTS . . . . . . . . . . .

Tests of the Theoretic Hypotheses . . . .
Tests of the Supplemental Hypotheses . . .
Supplemental Credibility Analysis . . . .
Experimental and Control Group Comparisons . .
Reliability of the Instruments . . . . .

IV DISCUSSION . . . . . . . . . . .

Summary of Procedures and Results . . . .
Implications for Future Research and Conclusions

APPENDIX

A COVER STORY AND EXPECTANCY INDUCTIONS . . . .

B EXPERIMENTAL MESSAGES . . . . . . . .

C MEASUREMENT INSTRUMENT . . . . . . . .

REFERENCES . . . . . . . . . . . . .

BIOGRAPHICAL SKETCH . . . . . . . . . .


Page

ii

iv














Abstract of Dissertation Presented to the Graduate Council
of the University of Florida
in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy


THE RELATIONSHIP OF VIOLATIONS OF RECEIVER EXPECTATIONS
AND RESISTANCE TO PERSUASION

By

Michael D. Miller

August 1978

Chairman: Michael Burgoon
Major Department: Speech

This investigation extended the development of a new model of

resistance to persuasion. The model views the induction of resistance

to persuasion an extension of the persuasion process. Support was

found for the prediction that violations of induced receiver expecta-

tions are mediators of resistance to subsequent persuasive attacks.

Positive violations of receiver expectations induce counterarguing

which will lead people to be resistant to a second persuasive message

advocating the same side of a given attitudinal issue. Negative

violations of receiver expectations decrease the probability of

counterarguing and increase the vulnerability of people to subsequent

persuasive attacks. Competing explanations for the results were ruled

out through the utilization of a counterbalanced design and stringent

control procedures.














CHAPTER I
RATIONALE


Of the research on resistance to persuasion, perhaps the most

consistently tested and supported hypotheses have been generated by

McGuire's (1964, 1969) inoculation model. McGuire bases the framework

of the model on a biological analogy:

. the person is typically made resistant to some
attacking virus by pre-exposure to a weakened dose of
the virus. This mild dose stimulates his defenses so
that he will be better able to overcome any massive
viral attack to which he is later exposed
(1964, p. 163)

McGuire suggests that analogous defensive techniques operate in

persuasive situations. Thus the primary emphasis of inoculation

research, and resistance to persuasion research in general, has

centered on preexposing subjects to weakened forms of persuasive

messages. This preexposure, or "inoculation," is operationalized

through the use of refutational messages which both introduce and

refute arguments against the subject's attitude that is to be made

resistant.

A relatively large body of research has indicated that various

pretreatment strategies are effective in inducing resistance to

persuasion. However, Burgoon, Cohen, Montgomery and Miller (1978)

have pointed out that this rather singular emphasis on inoculation

theory and Dretreatment message strategies has counterproductively

restricted hteory development in the area of resistance to persuasion.








Rather than viewing the induction of resistance to persuasion as

distinct from the persuasion process, Miller and Burgoon (1973)

suggest that the conferral of resistance to persuasion should be

conceptualized as an extension of the persuasion process. Burgoon

et al. (1978) indicate that such a conceptualization would allow for

a better integration of findings in the persuasion literature with

our present knowledge of the resistance to persuasion process, as well

as allowing more conceptual specificity.

Threat and Counterarguing

As indicated above, continued singular interest in testing

hypotheses directly drawn from the inoculation model may be less than

productive. However, the inoculation framework does advance several

theoretical notions which may be of heuristic value. One of the more

intriguing is the importance of threat to an individual's attitudes

in the conferral of resistance to persuasion.

The inoculation model assumes that threat is a necessary ante-

cedent to resistance to persuasion. It is presumed that any pretreat-

ment designed to make an individual's attitude resistant to subsequent

persuasive attacks must motivate the individual to defend that

attitude. This motivation to defend is supplied by exposing the

vulnerability of the attitude through the use of refutational pretreat-

ments. Refutational pretreatments are assumed to be intrinsically

threatening. However, supportive pretreatments (messages which Further

support or bolster the individual's existing attitudes) are assumed

to be inherently nonthreatening. This leads to the inoculation model

prediction that refutational defenses lead to greater resistance than

do supportive defenses (McGuire. 1964).








Since the inoculation model assumes the relative ineffectiveness

of supportive pretreatments is due to the lack of threat implied by

the message, it follows that an extrinsic threat to the attitude

combined with supportive information should increase the efficacy of

the pretreatment. Research (McGuire and Papageorgis, 1962) has

supported this reasoning.

Support for the importance of attitudinal threat has also been

found in numerous other studies. Tannenbaum (1967) points out that

one of the most impressive predictions of inoculation theory is that

the refutation of one set of arguments against a belief or attitude

generalizes to induce resistance to attack messages composed of novel

arguments (McGuire, 1961; Papageorgis and McGuire, 1961). The support

of this prediction is especially relevant to the assumed threat-

motivation-resistance relationship since the alternative explanation

that the superiority of the refutational pretreatment is a function of

remembering the rebuttals of the arguments is ruled out. McGuire

(1963) predicted and obtained results indicating that the addition of

an extrinsic threat increased the resistance-inducing effectiveness

of the pretreatments. Additionally, the effectiveness of the supportive

strategy was enhanced more than that of the refutational strategies.

In direct tests of the relationship between threat and resistance to

persuasion, Burgoon et al. (1976) and Miller and Burgoon (1977) have

found that threat is a predictor of resistance. Moreover, their

results indicate a curvilinear relationship such that moderate threat

leads to greater resistance than either low or high threat.

Taken together, the empirical evidence strongly supports the

notion that threat to an individual's attitude leads to a motivation








to defend that attitude which, in turn, leads to increased resistance

to persuasion. Of major theoretic importance, then, are the defense

mechanisms through which the increase in motivation is manifested.

Festinger and Maccoby (1964) and Osterhouse and Brock (1970) have

emphasized that when exposed to persuasive attacks, individuals

actively subvocalize counterarguments against the position advocated.

This subvocalization of counterarguments is presumed to create resis-

tance to the persuasive information the individual is being exposed to.

It seems reasonable, then, that the effectiveness of the threat

component of certain pretreatments is a function of increased counter-

arguing stimulated by the awareness that the individual's attitudes

are vulnerable to attack.

The interest in counterarguing as a mediator of attitude change

has led to lines of research focusing on the effects of inhibiting

counterarguing, primarily through the presentation of some form of

distraction concurrent with the persuasive attack message. Festinger

and Maccoby (1964) hypothesized that since counterarguing creates

resistance to persuasion, any form of distraction which interferes

with that process will lead to increased acceptance of persuasive

attempts. However, McGuire (1966) and others (Freedman and Sears,

1965; Haaland and Venkatesan, 1968) have pointed out that the hypothesis

that distraction should lead to increased acceptance of counter-

attitudinal information is directly opposed to predictions derived

from a learning theory approach. The learning theory approach suggests

that distractions should interfere with the learning of the persuasive

information. This failure to adequately learn the persuasive arguments

should lead to an inhibition of the attitude change process.








Research concerned with these alternative theoretic explanations

has produced mixed results. Festinger and Maccoby (1964) supported

their hypothesis that distraction should facilitate attitude change.

Viewing of a humorous, irrelevant film served as the distraction.

Rosenblatt (1966) used irrelevant slides as a distraction and also

found support for the distraction-increased acceptance hypothesis.

Freedman and Sears' (1965) results indicated a trend in the direction

of the distraction-increased acceptance hypothesis, but failed to

reach statistical significance. Osterhouse and Brock (1970) report

two studies, both indicating that counterargument production is

inhibited and communication acceptance is enhanced by distraction.

However, other studies have found an inverse relationship between

distraction and communication acceptance. Haaland and Venkatesen

(1963) found that the use of both irrelevant film presentations and

answering multiple choice and semantic differential items as dis-

tractors decreased attitude change. Vohs and Garrett (1963) also used

two different types of distractions (operations upon geometric shapes

and solving simple arithmetic problems) and found the inverse relation-

ship between distraction and attitude change. Still other studies

have failed to find any relationship between distraction and attitude

change (Breitrose, 1966; Gardner, 1966).

Several studies have investigated the interactive effects of

distraction and other variables upon attitude change. Kiesler and

Mathog (1968) found that the direct relationship between distraction

and communication acceptance held only when the source of the

communication was highly credible. They interpret this finding as

being in keeping with the distraction-increased acceptance hypothesis,








reasoning that if the speaker is low on credibility the need to

counterargue is diminished. The receiver can resist the persuasive

message by simply derogating the source of the persuasive message.

Other research (Regan and Cheng, 1973) indicates that the

structure of the persuasive attack message interacts with distraction.

When presented with a complex message, subjects in distraction condi-

tions changed their attitudes less than nondistracted subjects.

However, when the message was simple to comprehend, distracted subjects

exhibited more attitude change than did nondistracted subjects.

Although a great deal of research has indicated that various

types of distractions mediate attitude change in similar ways (see

Baron, Baron and Miller, 1973, for a review), the type of distraction

employed does seem to have an effect on the amount and direction of

attitude change obtained. Zimbardo, Snyder, Thomas, Gold and Gurwitz

(1970) found that "receiver set" mediated the effects of distraction.

When subjects were instructed to attend primarily to distracting

tasks, attitude change was decreased. However, when instructed to

attend primarily to the message instead of the distraction, distraction

increased attitude change. Using similar procedures and distraction

tasks, these results were replicated by Insko, Turnbull and Yandell

(1974).

In sum, it would seem that there is little doubt that a subvocal

counterarguing process does exist and that this phenomenon mediates

the attitude change process. Although the results from distraction

studies have been mixed, Osterhouse and Brock (1970) and Baron. Baron

and Miller (1973) have concluded that distraction during the presenta-

tion of an influence attempt increases the persuasive impact of that








message, unless the distraction is severe enough to interfere with the

cognitive processes of comprehension and recall of the persuasive

arguments. Thus it seems that the relationship between distraction

and communication acceptance is curvilinear, a relationship incorpo-

rating and compatible with both the suppression of counterarguing

hypothesis and with learning theory explanations.

However, it should be noted that none of the studies discussed

above investigated the effects of distraction during an initial

message on acceptance of subsequent persuasive messages on the same

topic. Conversely, the traditional resistance to persuasion paradigm

has been concerned with the effects of pretreatment messages on

acceptance of subsequent persuasive attacks, but has virtually ignored

the ability of individuals to adequately resist persuasive attacks

through the subvocal generation of counterarguments. Indeed, while

acknowledging the existence of the counterarguing process, most

inoculation research seems to assume that without being provided

specific defenses against persuasive attempts the individual will

change his attitudes in the direction advocated.

Threat and Violations of Expectations

Another line of research that has been somewhat overlooked in the

traditional resistance to persuasion literature concerns violations of

receiver expectations of communicative behavior. The majority of

findings that are discussed in the literature have consisted of post

hoc explanations for unexpected results in traditional persuasion

studies (Burgoon, Jones and Stewart, 1974).

In one of the earliest systematic investigations of the effects

of language intensity (the degree to which a source's language








expresses deviation from neutrality toward a concept) on attitude

change, Bowers (1963) predicted that highly intense language would

produce more attitude change than language of low intensity. The

results of the experiment, however, indicated a significant difference

opposite the direction predicted. That is, low intensity messages

produced more attitude change than did high intensity messages. One

explanation for the results is that the highly loaded statements

(comparing a type of educational system to prostitution, calling

American women "deranged," etc.) constituted a very severe negative

violation of the receivers' expectations of the communication

behavior of the source. This violation may have led to a "boomerang"

effect on attitude change.

Brooks and Scheidel (1968) unexpectedly found that an initially

unfavorably evaluated source was evaluated more positively after he

delivered a short prayer. Concern over whether this finding was

generalizable, or simply an artifact of the research, led Brooks

(1970) to conduct a more systematic investigation into the phenomenon.

His results indicated that the phenomenon is indeed generalizable.

Specifically, his research indicated that after exposure to a short

message, sources who were previously evaluated unfavorably were

evaluated more positively; and sources who were initially evaluated

favorably were evaluated more negatively. Brooks advanced a post hoc

explanation of the results concerning possible "contrast effects"

which may have been produced by violations of receivers' expectations:

. unfavorable (or favorable) speakers may be perceived
more (or less) favorably not because their behavior is
intrinsically persuasive (or dissuasive) but because
it contrasts with stereotyped expectations which audiences
hold for notorious (or popular) public figures. (1970,
p. 155)








In a related study, Burgoon (1970) preceded the presentation of

a moderate message on race relations with the introduction of a "name

set." Although the message was not attributed to the name mentioned

in the name set manipulation, it was predicted that response sets

introduced through the evaluation of "militant" names would lead to

the perception that the message was more militant than would response

sets introduced through the evaluation of nonmilitantt" names.

Contrary to the prediction, whites who were given the nonmilitant set

judged the message to be more militant than did those who received the

militant set. Burgoon (1970) suggested that since the white subjects

reacted very unfavorably toward the militant names, that the moderation

of the subsequent message introduced a contrast effect. Burgoon also

echoes Brooks (1970) by concluding that the reactions of individuals

to speakers may often be a function of the speaker's communication

behavior contrasting with stereotyped expectations.

One study has attempted to integrate the violations of expecta-

tions notion into the traditional resistance to persuasion paradigm.

Burgoon and Chase (1973) reasoned that exposure to refutational

pretreatments creates expectations about future messages which may be

used in influence attempts. They suggested that the most effective

refutational strategy would be one that matches the intensity of the

persuasive attack message. This strategy would avoid the contrast

effect that might lead to more attitude change when a highly intense

pretreatment message is followed by a message of moderate or low

intensity and would also avoid the possibility of a low intensity

pretreatment failing to threaten the subjects enough to motivate them

to defend their attitudes. Their results support their reasoning.








Although most of the above studies discussed their findings on a

post hoc basis, Burgoon, Jones and Stewart (1974) have attempted to

integrate the violations of expectations notion into a limited

propositional framework. Based on the research reviewed above they

advance two propositions concerning violations of expectations,

language intensity and attitude change:

Given the passive message reception condition, when a
source uses a level of language intensity that violates
the receiver's expectations in a positive manner, signifi-
cant attitude change will occur in the direction advocated
by the source.

Given the passive message reception condition, when a
source takes an unexpectedly intense position, it will
result in minimal or even negative attitude change.
(Burgoon, Jones and Stewart, p. 243)

In the first study designed to test this reasoning, an interaction

hypothesis was advanced such that female sources would be more

effective utilizing low intensity language, while male sources would

be least effective with low intensity language. Burgoon, Jones and

Stewart (1974) assumed that the stereotype of females in this society

involves such impressions as "submissive," "complimentary" and

"domestic." Based on the propositions discussed above, it was

reasoned that highly intense language would violate the receiver's

expectations of the communicative behavior of a female in a negative

manner, thus leading to reduced attitude change.

The second study hypothesized that low credible speakers would be

more persuasive using low intensity language, while high credible

speakers would be more effective utilizing highly intense language.

This hypothesis received partial support.








Taken together, the research discussed above and the propositional

framework advanced by Burgoon et al. (1974) strongly support the

notion that violations of receiver expectations are an important

mediator of the attitude change process. In addition, the research

of Burgoon and Chase (1973) has emphasized the importance of receiver

expectations in the traditional resistance to persuasion paradigm.

A New Model of Resistance to Persuasion

As indicated above, most research on resistance to persuasion

has maintained consistency with the inoculation approach and has

dealt almost exclusively with the development of pretreatment

message strategies designed to lessen the impact of a future persuasive

message. Although the inoculation model has served a valuable and

informative function, there is a limit to the heuristic capacity of

any analogy. Indeed, it has been argued that the continued singular

interest in the inoculation model has restricted rather than enhanced

our knowledge of the resistance to persuasion process.

Burgoon et al. (1978) have developed a new model of resistance

to persuasion that attempts to expand the restricted view of the

inoculation approach. While retaining the notion of attitudinal

threat as an important variable, they suggest that for conceptual

reasons it is best to view the resistance to persuasion process as

an extension of the persuasion process. Conceptually, this type of

approach allows for the integration of existing knowledge of persuasion

into our theorizing and research on resistance to persuasion.

This model attempts to integrate our current knowledge of the

effects of distraction and receiver expectations with McGuire's

notions on the importance of threat in the resistance to persuasion







process. They suggest that receiver expectations and attitudinal

threat can be affected by directing the attention of receivers to

different types of distracting tasks during the presentation of

persuasive message. Differential types of critical tasks are assumed

to induce differing critical response sets (or receiver expectations)

which, in turn, are related to the amount of counterargumentation

produced during and subsequent to the presentation of an initial

persuasive message. This differential in counterargumentation is

assumed to enhance or inhibit attitude change produced by subsequent

persuasive messages.

More specifically, this model suggests that when people are

asked to critically evaluate attributes of a communicator and are to

look for negative source and/or delivery characteristics, they should

be vulnerable to subsequent persuasive attacks. Burgoon et al. (1973)

argue that several factors combine to produce this effect. The

evaluation of negative speaker characteristics should inhibit the

persuasive effectiveness of the message and should minimize threat to

the receiver. In addition, the attention to source characteristics

should distract the person from counterarguing crucial arguments in

the persuasive message. Thus after the reception of one persuasive

message, the receiver will be unpracticed in defending his or her

attitudes and, because of the lack of attitudinal threat, unmotivated

to counterargue in the future. This should result in increased

vulnerability to a second persuasive attack.

Conversely, attention to positive source characteristics should

enhance the effectiveness of an initial persuasive message. Also.

threat to the receiver should be increased since his or her attitudes








have been shown to be vulnerable. The increase in motivation induced

by the attitudinal threat results in increased counterargumentation

which should shift attitudes back toward the initial negative position

after receipt of a second message. Research has supported both of

these predictions.

Although their initial research incorporated notions of receiver

expectations, Burgoon et al. (1978) do not explicitly deal with the

importance of confirmation or disconfirmation of these expectations.

They assume that the expectations themselves affect the amount of

threat to the receiver. Although their research seems to support this

view, it can be argued that an equally important factor is whether

these receiver expectations are confirmed or, if disconfirmed, what

form of disconfirmation is involved. As previous research (Bowers,

1963; Brooks, 1970; Brooks and Scheidel, 1968; Burgoon, 1970; Burgoon

and Chase, 1973; Burgoon, Jones and Stewart, 1974) indicates, viola-

tions of receiver expectations lead to powerful effects on message

reception and attitude change.

Consistency between the previous research and the propositional

framework advanced by Burgoon et al. (1974) demonstrates that when a

source violates a receiver's expectations in a positive manner, the

effect of an initial message should be enhanced. Similarly, when a

source violates receiver expectations in a negative manner, the effect

of an initial message is inhibited. It would seem valuable to

integrate this knowledge with the new model of resistance to persuasion

developed by Burgoon et al. (1978).

It can be argued that when an individual is asked to evaluate the

highly intense characteristics of a counterattitudinal message, and the








message consists entirely of low intensity language, that the

individual's expectations are violated in a positive manner. This

reasoning is consistent with the theorizing and research of Burgoon,

Jones and Stewart (1974). As indicated above, this type of positive

violation should lead to a contrast effect which should enhance

attitude change after the receipt of this first message. However, the

realization that persuasive arguments counter to one's beliefs can be

advanced in a reasonable, moderate manner should threaten the individ-

ual and increase the motivation to counterargue in the future. Thus

upon exposure to a second persuasive message on the same attitudinal

issue, attitude change should be inhibited, or even reversed.

Similarly, when an individual is asked to evaluate the low

intense statements of a counterattitudinal message, and the message

consists entirely of highly intense statements, the individual's

expectations are violated in a negative manner. This negative viola-

tion should prove distracting and should also inhibit the counter-

argumentation process. However, reasoning analogous to that of

Kiesler and Mathog (1968) indicates that in this type of situation,

the need to counterargue is reduced. Persuasive information can be

resisted through derogation of the source or of the message style

without the need to systematically counter each argument. This

ability to resist without counterargumentation should decrease threat

to the individual and reduce the motivation to counterargue in the

future, thus leading to increased vulnerability to subsequent persua-

sive message on the same attitudinal topic.

It is evident that this line of reasoning is consistent with the

prior research on violations of receiver expectations and is also








compatible with the model of resistance to persuasion advanced by

Burgoon et al. (1978). It maintains consistency with their view that

any message may affect the persuasive efficacy of a subsequent persua-

sive attack. By suggesting that a critical variable in this process

is the confirmation or disconfirmation of the expectations developed

by the induced response sets, the reasoning advanced above also

extends their assumption that the induction of critical response sets

may either inhibit or enhance the effect of forthcoming messages.

Derivation of the Hypotheses

The model of resistance to persuasion advanced by Burgoon et al.

(1978) provides a synthesis of research illustrating the relationship

of threat, distraction, counterargumentation and critical response

sets to resistance to persuasion. It can be argued that the confirma-

tion or disconfirmation of these critical response sets are of over-

riding importance in this view of resistance to persuasion. Positive

violations of receiver expectations create a contrast effect leading

to an enhancement of attitude change attributable to a first persuasive

message. This enhancement effect points out the vulnerability of the

receiver's attitudes, thus threatening him and motivating him to

counterargue in the future. The increased counterargumentation will

lead to a reversal of attitude change upon exposure to a second

persuasive message on the same attitudinal topic. This reasoning

leads to the first hypothesis:

H1: People who are initially induced to expect highly

intense messages of a counterattitudinal nature,

but who receive low intensity messages, will be

initially more positive toward the attitude issue.








Upon receipt of a second attack message, however,

these individuals will revert to being more negative.

In an analogous manner, when an individual's expectations are

violated in a negative manner, the effect of a first persuasive

message will be inhibited. However, the individual will not feel

threat to his attitudes and thus will not be motivated to counter-

argue in the future. Upon receipt of a second persuasive message on

the same topic the individual will not be adequately prepared to

defend his attitudes. This will result in attitude change in the

direction advocated by the persuasive message.

H2: People who are initially induced to expect low

intensity messages, but who receive highly intense

messages, will be initially more negative toward

the attitudinal issue, but will demonstrate attitude

change in the direction advocated by a second

persuasive attack.

Both Hypothesis 1 and Hypothesis 2 deal with violations (or

disconfirmations) of receiver expectations of the communication

behavior of a source. Although no specific predictions will be

formalized, it is important to note that the rationale advanced by

this investigation assumes differences in the outcomes produced by

disconfirmations and confirmations of receiver expectations.

Based on the reasoning and prior research of Burgoon et al.

(1978), a second set of hypotheses may be advanced. Exposure to a

first persuasive message should be perceived as more threatening than

a subsequent message received after time has been allowed to prepare

defenses. Individuals should indicate more confidence in their








decisions after exposure to two messages since they will have prepared

defenses and already have changed or maintained their attitudes. The

second attack message should be perceived as more expected than the

first persuasive attack since subjects will have already seen the

same attitudinal position advocated in the prior communication.

H3: There will be a main effect for perceived threat

with threat decreasing after receipt of a second

persuasive message.

H4: People will express more confidence in their attitudes

toward the message topic after receipt of the second

message.

H : Even though both messages argue the same side of

an attitude issue, people will find the position

advocated in the second message as more expected.

The rationale leading to the first two hypotheses assumes that

attitudinal threat induced by exposure to a first persuasive message

will motivate people to counterargue against future persuasive attacks.

This assumption suggests that threat produced by exposure to a first

persuasive message should be negatively related to attitude scores

after a second persuasive message. This reasoning leads to a final

hypothesis:

H6: There will be a significant negative relationship

between attitudinal threat after exposure to a first

persuasive attack and attitudes toward the issue

after receipt of a second persuasive attack.

Previous research (Burgoon et al., 1978) has indicated that

violations of expected language usage may lead to differing perceptions








of communicator credibility. As they indicate, however, there is no

theoretical base upon which to base exact predictions. Thus the

following research questions are advanced:

Q1 To what extent will confirmations and disconfirmations

of expected message intensity affect receiver per-

ceptions of a source's competence, composure,

character, sociability and extroversion?

Q2: To what extent will confirmations and disconfirmations

of response sets induced prior to a first persuasive

attack affect perceptions of competence, composure,

character, sociability and extroversion of a source

delivering a second persuasive message on the same

attitudinal topic?














CHAPTER II
METHODS


Design of the Experiment

Subjects were 219 undergraduate students enrolled in basic

communication courses at the University of Wyoming. Subjects were

randomly assigned to eight control groups (n=144) and four experimental

groups (n=75). Each subject in the experimental conditions read two

persuasive messages advocating the acceptance of a counterattitudinal

position. Each message argued for the acceptance of the position on

the basis of a different set of issues. Before presentation of the

first message, experimental subjects were randomly assigned to one of

two expectancy sets. One set induced the subject to evaluate the low

intensity statements in the persuasive message, while the other induced

the subject to focus on the highly intense statements of the message.

Subsequently, each experimental subject was randomly exposed to either

a high or low intensity persuasive message. At a later experimental

session, subjects were exposed to a moderately intense second persua-

sive message advocating the same counterattitudinal position. No

expectancy sets were experimentally induced prior to the presentation

of the second message. Message order effects were controlled by

randomly varying the order of message topic presentation within each

experimental group ana within the two message control groups.

Adequate tests of the hypotheses mandated the use of several

control groups in the experiment. Prior to the presentation of a







low intense persuasive message, one control group received instructions

to focus on both the high and low intensity language of the message.

This control received a second message (moderate intensity) at a

subsequent experimental session. A second control group received

identical instructions and followed identical procedures, with the

exception that the first experimental message they received was highly

intense. The third control group received no expectancy set, but was

presented with a low intensity message followed at a later experimental

session by a moderate intensity message. Similarly, a fourth control

group received no expectancy instructions, but was presented with a

highly intense message followed by a moderately intense message at the

second experimental session. The fifth control group received no

expectancy instructions and received only one message (highly intense).

The sixth control group also received only one message (low intensity)

and received no instructions designed to introduce an expectancy set.

The seventh control group received only one message (moderate inten-

sity) and received no instructions to induce an expectancy set. The

final control group received only a pretest and a posttest on their

attitudes toward the experimental topic.

The utilization of the offset control group follows a recent trend

in the research on resistance to persuasion. Burgoon and King (1974)

have pointed out two distinct advantages of this design. First, it

allows the determination of pretest attitudes from a random sample of

the subjects. Secondly, and of critical importance to research on

resistance to persuasion, it circumvents the problem of pretest

sensitization encountered in change score designs. Since the pretest

itself may induce resistance to persuasion, this design was crucial to

the investigation. Table 1 illustrates the design.













ra

,0C U U U U U, aUJ
r E o o 0 0 Ca a a c uCa Ca cu oo
- z Z > > >- >- >- >
S o
ca Ca 4--i
a-a
I LWU,



-- m-
LQIV-








SL 0 0 0 0- 0 I
C I- l J Ca .J = 1 -
c)I Ca
fQ 4-)o m cu L

(n0 A Ca cu
I Ea

0
,- 04-




ro S-
r ,--U
SI E








E0 0
S* ro

roE 0 ) Ca c (C QC CaJ Ca Ca a CL
U *- Ca Ca

Sro









0 0
C C C a a a










I o -
S.. o = -
u c- Ii















S -, o a E -- < "
r C C = 3 > 0 C U0a







U 0 0 0 0 0 0 0 a -













-- U- U- u-, U 0 2 E 2 2
(. Ca Ca a 0 a a a -- -- -- I
a Z 0 0 -1 U co
Q. -j -J r" '1
SC I 0

m C3 U *X











J Ca-
3 0 0


0-) 0- 0-1 0- 1- i- i Lii -L
.-I-0

L.) 0 .0- 0
S0- 0- 0 0 a 0 I 0 I0



CL o o o t oro o I a a 4-| =a













M mc LU o








Message Construction

The experimental messages utilized in the Burgoon et al. (1978)

study served as the basis for the experimental materials used in the

present investigation. Burgoon et al. point out that a significant

body of research has indicated that attitude change is mediated through

message variables. Thus in that investigation, as well as in the

present study, it was deemed critical that certain message variables be

controlled and accounted for in the experimental messages. Failure to

insure these controls could render the results of the investigations

uninterpretable.

The two experimental messages developed for the Burgoon et al.

(1978) study each consisted of four arguments supporting the legaliza-

tion of heroin sales in the United States. One message argued on the

basis of health-related issues; the other argued on the basis of

crime-related issues. Since comprehension may be related to sentence

length, the two messages were developed to contain approximately the

same average sentence length. In addition, the messages were both

shown to have an extremely high Index of Contingency (Becker, Bavelas

and Braden; 1961), which is a sensitive index of comprehension.

Research by Clark and Begun (1971) has indicated that subject-predicate

compatibility results in more favorable evaluations of messages by

receivers. Thus the messages used by Burgoon et al. (1978) used

stringent subjective controls in matching the messages on abstractness,

humanness and animation.

In addition, Burgoon et al. controlled for the effects of language

intensity by constructing both messages to be moderately intense.

Although part of this control was judgmental, some objective controls








were utilized. Highly intense metaphors (Bowers, 1964) were not used.

Present tense verbs were preferred over future tense verbs (McEwen,

1969); intense levels of adverbial qualification (Burgoon and Miller,

1971) and obscure or infrequently used words (Bowers, 1964) were

deleted.

The experimental messages of the Burgoon et al. study served as

the moderate intensity messages in the present investigation. High

intensity and low intensity versions of the messages were created by

systematically varying the adverbial qualification within the messages

and by varying the severity of negative outcomes associated with

failure to accept the positions advocated by the messages (Burgoon et

al., 1976; Miller and Burgoon, 1977). This procedure resulted in the

development of six experimental messages: high, moderate and low

intensity versions of both the health-related and crime-related

messages.

The message variables controlled in the Burgoon et al. investiga-

tion were similarly accounted for in this experiment. Each message

contained four arguments advocating the legalization of heroin sales

in the United States. The high, moderate and low intensity versions

of the crime-related message contained 593 total words (average

sentence length = 22.81 words), 526 total words (A.S.L. = 20.23 words),

and 585 total words (A.S.L. = 22.5 words), respectively. The high,

moderate and low versions of the health-related message contained 637

total words (A.S.L. = 23.59 words), 615 total words (A.S.L. = 22.78

words) and 631 total words (A.S.L. = 23.37 words), respectively.








Experimental Procedures

Male experimenters were randomly assigned to administer treat-

ments. Experimental packets were randomly distributed to subjects.

The first page of the experimental packets contained written instruc-

tions assigning subjects to experimental and control conditions prior

to exposure to the first experimental message. Experimental groups

then read a message which utilized either health- or crime-related

issues. Message issues were randomly assigned to groups of subjects.

Subjects evaluated the intensity of statements in the messages by

underlining phrases as they read. After reading the messages, each

subject completed a four-item attitude measure consisting of seven-

interval semantic differential-type scales bounded by the bipolar

adjectives good-bad, foolish-wise, pleasant-unpleasant, and worthless-

valuable. To test the research questions advanced by the study, each

subject also responded to the measures of five dimensions of source

credibility developed by McCroskey, Jensen and Todd (1972). In

addition, perceptions of threat, confidence in the decision, and

message expectancy were measured by three separate five-item scales

consisting of seven-interval semantic differential-type items bounded

by bipolar adjectives.

Experimental and control conditions receiving a second experi-

mental message were tested two to three days after the initial

experimental sessions. Subjects were instructed to read the messages

carefully. Nlo expectancy sets were introduced. After reading the

messages, subjects again completed the attitude, source credibility,

threat, confidence and expectancy measures. Control groups who

received only one message completed a differing form of the attitude





25


measure at the second experimental session. The pretest-posttest

control group filled out differing forms of the attitude measure at

both experimental sessions.













CHAPTER III
RESULTS


Tests of the Theoretic Hypotheses

Hypothesis 1 predicted that people induced to expect a highly

intense counterattitudinal message would evidence relatively more

positive attitudes toward the issue after exposure to a low intensity

message. Upon receipt of a second counterattitudinal message, however,

it was predicted that they would revert to a more negative attitude.

The second hypothesis predicted that persons whose expectations were

violated by receiving a high intense message when a low intense message

was expected would be initially more negative toward the attitude

issue but would demonstrate more positive attitudes upon receipt of a

second persuasive attack.

A 2 (expectancy of high or low message intensity) x 2 (receipt of

high or low intensity message) x 2 (time of measurement) repeated

measures analysis of variance with time of measurement as the repeated

variable was computed. The results of the analysis indicated signifi-

cant main effects for message expectancy (F = 9.41, df = 1/71;

p < .05), message intensity (F = 4.27, df = 1/71; p < .05), time of

measurement (F = 8.68, df = 1/71; p < .05), and a significant expect-

ancy by intensity interaction (F = 15.28, df = 1/71; p < .05). These

effects were overridden by a significant expectancy by intensity by

time of measurement interaction (F = 7.63, df = 1/71; p < .05).








Summaries of the cell means and the analysis of variance are presented

in Table 2.


TABLE 2

Iean Attitude Scores and Analysis of Variance
Summary for the Experimental Groups


Expected
Intensity


High


Low


Source of Variance


Time of
Time 1
High Int. Low Int.

6.53* 10.55
n=17 n=22

8.68 13.18
n=19 n=17

Analysis


Measurement
Time 2
High Int. Low Int.

10.47 7.68
n=17 n=22

11.73 15.06
n=19 n=17

of Variance Summary


Mean 16308.03 1
Expectancy 417.98 1
Intensity 189.47 I
EXP x INT 100.57 1
Subj. w. groups 3152.77 71
[error (between)]

Time 83.78 1
Time x EXP 34.49 1
Time : INT 147.41 1
Time x EXP : INT 73.58 1
Time x Subj w. groups 685.12 71
[error (within)]

*Range of attitude measure: 4 (opposed)


16308.03
417.98
189.47
100.57
44.41


83.78
34.49
147.41
73.58
9.65


367.26
9.41
4.27
2.26



8.68
3.57
16.28
7.63


to 28 (favorable)


A direct test of Hypothesis 1 was made by computing a correlated

t-test between attitudes toward the issue at Time 1 and Time 2. People

who expected a highly intense message, yet received a low intensity

message, indicated significantly more negative attitudes toward the

legalization of heroin after receipt of the second persuasive message

advocating that policy (t = 11.94, df = 21; p < .05). The combined


<.05
<.05
<.05
NSD



<.05
NISD
<.05
<.05







results of the analysis of variance and the a priori comparison provide

strong support for the first hypothesis.

The second hypothesis was also put to direct test by computing a

correlated t-test between Time 1 and Time 2. The results indicated

that those people expecting a low intense message, but receiving a high

intense message, had significantly more positive attitudes after

receipt of a second message (t = 11.74, df = 18; p < .05). The com-

bined results of the analyses provide strong support for Hypothesis 2.

The obtained three-way interaction and the pattern of the cell means

also indicate that disconfirmation of receiver expectations produces

different effects than do confirmations of receiver expectations.

Tests of the Supplemental Hypotheses

Hypotheses 3, 4 and 5 predicted main effects for time of measure-

ment on the dependent variables of threat, confidence in the expressed

attitudes and message expectancy. All groups who received induced

expectancy sets and both persuasive messages were used in the computa-

tion of three separate 3 x 2 x 2 repeated measures analyses of

variance, with time of measurement as the repeated variable in each

test. A summary of the cell means for each dependent variable is

presented in Table 3.

Hypothesis 3 predicted that less perceived threat would be

evidenced after the receipt of the second persuasive message. No

significant effects were indicated by the analysis, however. The

only effect to approach significance was the three-way interaction

(F = 2.49, df = 2/94; .05 < p < .10).

As predicted by Hypothesis 4, people expressed increased confi-

dence in their attitudes following the receipt of a second attack
















TABLE 3

Mean Scores of Threat, Confidence and Expectancy


THREAT


High
Low
Both


CONFIDENCE


High
Low
Both


Time 1
Intensity
High Low


22.88*
24.84
25.38


27.24*
25.68
25.17


23.40
23.11
21.25


28.00
26.08
25.33


Time 2
Intensity
High Low


22.29
21.32
23.17


25.76
26.68
28.42


23.23
24.06
24.69


29.20
27.38
29.33


EXPECTANCY


Hiqh
Low
Both


21.12*
18.47
18.42


20.68
16.94
18.15


21.64
20.37
18.58


19.91
17.71
19.92


*Range of scores on each measure: 5 (Low) to 35 (High)


message (F = 5.21, df = 1/90; p < .05). Analysis of variance summaries

for the dependent variables threat and confidence are presented in


Table 4.








TABLE 4

Analysis of Variance Summary for the Dependent
Variables Threat and Confidence


THREAT


Source of Variance

Mean
Expectancy
Intensity
EXP x INT
Subj. w. groups
[error (between)]

Time
Time x EXP
Time x INT
Time x EXP x INT
Time x Subj. w. groups
[error (within)]

CONF IDENCE

Source of Variance

Mean
Expectancy
Intensity
EXP x INT
Subj. w. groups
[error (between)]

Time
Time x EXP
Time x IlNT
Time x EXP x INT
Time x Subj. w. groups
[error (within)]

*Four subjects did not
the analysis.


103990.00
14.20
88.09
45.93
6028.09


6.04
26.89
6.82
96.09
1811.36


135379.25
40.06
52.34
27.14
3986.53


103990.00
7.10
88.90
22.97
64.13


6.04
13.45
6.82
48.05
19.27


1
2
1
2
90*


111.04
110.14
17.89
13.41
1918.22


135379.25
20.03
52.34
13.57
44.29


111.04
55.07
17.89
6.70
21.31


complete this scale and hence were not used in


1621.59
.11
1.37
.36


.31
.70
.35
2.49


3056.33
.45
1.18
.30


5.21
2.58
.84
.31


<.05
NSD
'ISD
NSD


NSD
PISD
NSD
<.10


<.05
NSD
NSD
NSD


<.05
NSD
NSD
NISD








Hypothesis 5 predicted that people would perceive the second

persuasive attack as advocating a position more expected than the first

persuasive attack. The failure of any effects to approach significance

lends no support to this hypothesis. A summary of the analysis of

variance is presented in Table 5.


Source of Varian

Mean
Expectancy
Intensity
EXP x INT
Subj. w. groups
[error betweene

Time
Time x EXP
Time x INT
Time x EXP x INT
Time x Subj. w.
[error withinr


TABLE 5

Analysis of Variance Summary for the
Dependent Variable Expectancy

ice SS df MS

71527.38 1 71527.33
254.39 2 127.19
37.24 1 37.24
51.20 2 25.60
7096.75 94 75.50


_n)]





groups
)]


25.19
20.93
.92
13.66
1565.45


25.19
10.48
.92
9.33
16.65


Hypothesis 6 predicted a negative relationship between threat at

Time 1 and attitudes toward the issue at Time 2. A Pearson Product

Moment Correlation indicated a significant negative correlation between

these variables (r = -.23, p < .05), thus providing support for the

hypothesis.

Supplemental Credibility Analysis

The research questions concerned the effects of confirmation and

disconfirmation of receiver expectations of message intensity on the

credibility of the source of a persuasive message and on the credibility


F

947.42
1.68
.49
.34



1.51
.63
.05
.56


p
<.05
NSD
NSD
NrSD



NSD
NSD
NSD
NSD








of a source delivering a second persuasive message on the same

attitudinal topic. To test these questions all subjects who had

received expectancy sets and who had received two persuasive attack

messages were utilized in computing five successive 3 x 2 x 2 repeated

measures analyses of variance, with time of measurement as the

repeated variable. Table 6 presents a summary of the cell means for

each of the dimensions of credibility used in the analysis.

Competence, Character and Composure

Significant time of measurement main effects were indicated for

the dependent variables of competence, character and composure

(Competence: F = 11.56, df = 1/90; p < .05; Character: F = 8.49,

df = 1/90; p < .05; Composure: F = 5.11, df = 1/90; p < .05).

Analysis of variance summaries for these variables are presented in

Table 7. The effect was such that the source of a second attack

message was perceived as more competent, of higher character and more

composed. Since, however, this effect is evident in the distraction

control cells as well as in the cells that received confirming or

disconfirming messages, the effect cannot be attributed to the

manipulated independent variables. No other main effects or inter-

actions obtained significance.

Sociability and Extroversion

No main effects or interactions approached significance for the

dependent variables of sociability or extroversion. Thus confirmation

or disconfirmation of expected message intensity had no effect on the

perceptions of initial source credibility or on the perceptions of the

credibility of the source of a subsequent persuasive message on the

same attitudinal topic. Analysis of variance summaries are presented

in Table 8.








TABLE 6

Means on Credibility Dimension for Experiment
and Distraction Control Groups


COMPETENCE


High Expectancy
Low Expectancy
Both


High Int. Low Int.
12.71* 12.25
12.42 12.62
13.00 10.80
Time 1


CHARACTER


High Expectancy
Low Expectancy
Both


High Int.
12.53*
11.59
11.83


Time 1


Low Int.
11.25
12.15
11.80


High Int. Low Int.
13.94 12.50
12.53 13.46
13.50 11.87
Time 2


COMPOSURE


High Expectancy
Low Expectancy
Both


High Int.
13.06*
12.89
12.58


Time 1


SOCIABILITY


High Expectancy
Low Expectancy
Both


High Int.
14.47*
12.74
13.50


Time 1


Low Int.
13.20
12.85
12.53


High Int. Low Int.
13.24 12.55
12.68 13.00
12.83 12.87
Time 2


EXTROVERSION


High Expectancy
Low Expectancy
Both


High Int.
14.88*
15.63
17.25


*Range of scores on each measure: 3 (Low) to 21 (High)


Hiah Int.
13.12
13.68
13.25


Low Int.
13.15
15.08
13.00


Time 2


Low Int.
12.20
13.38
11.13


High Int.
13.06
13.68
13.00


Low Int.
13.05
14.23
13.07


Time 2


Low Int.
16.50
14.85
16.87


Time 1


High Int.
15.53
15.16
15.75


Low Int.
16.70
15.31
16.33


Time 2







TABLE 7

Analysis of Variance Summaries for the Dependent Variables
Competence, Character and Composure

COMPETENCE

Source of Variance SS df MS F R

Mean 30957.85 1 30957.85 1550.29 <.05
Expectancy 27.09 2 13.54 .68 NSD
Intensity 2.13 1 2.13 .11 NSD
EXP x INT 29.20 2 14.60 .73 NSD
Subj. w. groups 1797.22 90* 19.97
[error (between)]

Time 72.15 1 72.15 11.56 <.05
Time x EXP 12.22 2 6.11 .98 NSD
Time x INT 17.03 1 17.03 2.73 NSD
Time x EXP x INT 4.15 2 2.08 .33 NSD
Time x Subj. w. groups 561.74 90 6.24
[error (within)]

CHARACTER

Source of Variance SS df MS F R

Mean 28558.66 1 28558.66 1780.70 <.05
Expectancy 2.88 2 1.44 .09 NSD
Intensity 10.66 1 10.66 .66 NSD
EXP x INT 39.39 2 19.69 1.23 NSD
Subj. w. groups 1443.41 90* 16.04
[error (between)]

Time 56.93 1 56.93 8.49 <.05
Time x EXP 1.67 2 .83 .12 NSD
Time x INT 2.53 1 2.53 .38 NSD
Time < EXP x INT 7.29 2 3.65 .54 NSD
Time x Subj. w. groups 603.58 90 6.71
[error (within)]

COMPOSURE

Source of Variance SS df MS F R

Mean 31067.38 1 31067.38 2677.57 <.05
Expectancy 36.52 2 18.26 1.57 NSD
Intensity 1.90 1 1.90 .16 NSD
EXP x INT 12.19 2 6.10 .53 NSD
Subj. w. groups 1044.26 90* 11.60
[error (between)]








Table 7 continued


Source of Variance

Time
Time x EXP
Time x INIT
Time x EXP x INIT
Time x Subj. w. groups
[error (within)]

*Four subjects did not
in the analysis


SS df


30.10
4.41
7.56
3.85
530.62


complete these measures, hence were not used


TABLE 8


Analysis of Variance Summaries for the Dependent Variables
Sociability and Extroversion


SOCIABILITY


Source of Variance

Mean
Expectancy
Intensity
EXP x INIT
Subj. w. groups
[error (between)]

Time
Time x EXP
Time x IIT
Time < EXP x INT
Time x Subj. w. groups
[error (within)]

EXTROVERSION

Source of Variance

Mean
Expectancy
Intensity
EXP x INT
Subj. w. groups
[error (between)]


SS df


31513.57
11.30
7.82
11.90
1038.92


5.77
9.27
4.13
1.13
360.12


1
2
1
2
90*


1
2
1
2
90


SS df


46844.93
49.57
7.13
27.10
1351.27


1
2
1
2
90*


MS

30.10
2.20
7.56
1.93
5.90


5.11
.37
1.28
.33


<.05
NSD
NSD
NSD


MS

31513.57
5.65
7.82
5.95
11.54


5.77
4.64
4.13
.56
4.00


F

2729.96
.49
.68
.52



1.44
1.16
1.03
.14


p

<.05
ISD
NSD
NSD



NISD
NSO
NSD
NSD


46844.93
24.78
7.13
13.55
15.01


3120.05
1.65
.47
.90


<.05
NSD
NSD
NSD








Table 8 continued


Source of Variance SS df MS F p

Time 1.35 1 1.85 .36 NSD
Time x EXP 16.35 2 8.17 1.60 NSD
Time x INT 2.72 1 2.72 .53 NSD
Time x EXP x INT 5.47 2 2.73 .54 NSD
Time x Subj. w. groups 458.89 90 5.10
[error (within)]

*Four subjects did not complete these measures, hence were not used
in the analysis.




Experimental and Control Group Comparisons

An analysis using Duncan's multiple range test was computed to

compare experimental and control conditions at both Time 1 and Time 2.

Table 9 presents a summary of the means and results of the multiple

comparisons.

At Time 1, only the experimental groups receiving low intensity

messages that confirmed or disconfirmed receiver expectations differed

significantly from the pretest-posttest only control condition.

Although no predictions concerning confirmation of expectations were

advanced, this is consistent with the Hypothesis 1 prediction that the

disconfirmation of expected high intensity messages would immediately

enhance attitudes toward the experimental issue. No control group

differed significantly from the pretest-posttest only control

condition.

At Time 2, the experimental group receiving disconfirmation of

expected high intensity messages did not differ significantly from

the pretest-posttest only control group, even though differences did

exist at Time 1. This finding is also consistent with Hypothesis 1.









TABLE 9

Experimental and Control Gr


Group Time 1 Attitu

Pretest-Posttest Only 6.53

One Message Control 7.06
(High Intensity)

One Message Control 9.88
(Moderate Intensity)

One Message Control 7.81
(Low Intensity)

Two Message Control 9.18
(High Intensity)

Two Message Control 8.09
(Low Intensity)

Distraction Control 7.87
(Low Intensity)

Distraction Control 7.58
(High Intensity)

Experimental I 13.18'
(Low EXP, Low INT)

Experimental II 8.68
(Low EXP, High INT)

Experimental III 10.55
(High EXP, Low INT)

Experimental IV 6.53
(High EXP, High INT)


*Indicates significant difference from
(p < .05) at Time 1.

**Indicates significant difference from
(p < .05) at Time 2.


"oup Comparisons


ide Mean Time


2 Attitude Mean

6.37

7.24


9.81


8.54


11.71**


10.60**


9.40


10.58


15.06**


11.74**


7.68


10.47


pretest-posttest group


pretest-posttest group








The persuasive effectiveness of the combination of two messages is

demonstrated by the significant differences between the pretest-

posttest only condition and each of the two-message only control

conditions. As predicted by Hypothesis 2, the experimental group

experiencing disconfirmation of expected low intense language behavior

differed significantly from the pretest-posttest only control group,

even though this difference was not apparent at Time 1. The group

experiencing confirmation of expectations of low intensity message

reception remained significantly different from the pretest-posttest

only control group at Time 2. Taken together, the experimental-

control group comparisons lend further support to the rationale leading

to Hypothesis 1 and Hypothesis 2.

Further comparisons indicated no significant change in the

pretest-posttest only control group between Time 1 and Time 2.

Similarly, no significant change occurred in any of the one-message

only control groups between Time 1 and Time 2. Thus any differences

between Time 1 and Time 2 in the other control and experimental groups

may be attributed to the experimental manipulations. A comparison of

the health-related and crime-related messages indicated no differences

in effectiveness, thus obviating the need to consider this factor in

the analysis.

Reliability of the Instruments

As a check on the internal consistency of the measurement instru-

ments utilized in the study, Cronbach's coefficient alpha was computed

for each dependent measure. Table 10 presents a summary of the

coefficients for each instrument.








TABLE 10

Reliability Coefficients for the Dependent Measures


Measure :

Attitudes .93

Threat .78

Expectancy .76

Confidence .79

Competence .80

Character .63

Composure .72

Sociability .52

Extroversion .82





The reliability coefficient for the instrument measuring the

variable of primary theoretic interest (attitudes toward the legaliza-

tion of heroin) was extremely high (cc = .93). The measures of threat,

expectancy and confidence in the decision were also satisfactory

(ca > .75). The consistency of the measures of perceived source

credibility were more variable, ranging from a low of .52 (for

sociability) to a high of .82 (for extroversion).














CHAPTER IV
DISCUSSION


Summary of Procedures and Results

This investigation tested two theoretic hypotheses concerning the

relationship of violations of receiver expectations to resistance to

subsequent persuasive attacks. The study utilized 219 subjects

randomly divided into four experimental and eight control groups.

Since the rationale of this study is an extension of the reasoning

advanced by Burgoon et al. (1978), an effort was made to facilitate

between-experiment comparisons by using the same attitudinal issue as

the experimental topic in both studies. The experimental issue,

proposed legalization of the sale of heroin in the United States, was

initially chosen because it is highly counterattitudinal for the

subjects in the experiment.

Since the hypotheses concerned the effects of initial violations

of receiver expectations on acceptance to subsequent persuasive

attacks, each experimental subject was exposed to two persuasive

messages. This required the construction of two messages which used

different arguments to advocate the legalization of heroin sales in the

United States. The persuasive messages used in the Burgoon et al.

(1978) study served as the basis of the experimental messages used in

this investigation. One message argued for the legalization of heroin

sales on the basis of lowering the crime rate, while the other message

argued on health-related issues. Since the intensity of the first








persuasive attack was a manipulated variable, three versions of each

message were prepared. Care was taken to rule out potential inter-

vening effects of other language and message variables.

Prior to exposure to the first persuasive message, each subject

in the experimental conditions was led to expect either a high or a

low intensity message. No information was given as to the persuasive

nature of the messages or of the message topic. Each subject then read

either a high or low intensity message advocating legalization of

heroin sales. Two to three days later experimental subjects were

exposed to a second persuasive message. All subjects received a

moderately intense message using arguments different from those

previously read. Message topic and message order differences were

controlled by randomly varying the order of presentation within each

experimental group. As an additional check, the effect of message

topic on attitude scores was compared. No significant differences

were found. Thus differences between attitude scores of the experi-

mental groups cannot be attributed to message differences.

In addition to these procedures, several control groups were also

utilized. The effects of message intensity and the presentation of

only one persuasive message were assessed by the use of three one-

message only control groups. One of these groups received a high

intensity message; one received a low intensity message; and one

received a moderate intensity message. The effect of receiving two

persuasive messages was assessed through the use of two separate two-

message only control groups. One group received a high intensity

message followed by a moderate intensity message, while the other

group received a low intensity message followed by a moderate intensity







message. Simple distraction effects were checked through the

utilization of two separate two-message distraction control groups.

One of these groups received a high intensity message followed at a

later date by a moderate intensity message, while the other group

first read a low intensity message, followed by a moderate intensity

message. Both groups received instructions to evaluate both the high

and low intensity statements in the first persuasive message. While

this task was distracting, it did not induce expectations which would

have been confirmed or disconfirmed as in the experimental conditions.

A final control group received only a pretest and a posttest on

attitudes toward the legalization of heroin sales.

The theoretic hypotheses were tested with a three by two by two

repeated measures analysis of variance. The independent variables

were receiver expectations concerning the first persuasive message

(either high or low intensity), actual intensity of the first

persuasive message (either high or low intensity) and time of measure-

ment (either immediately following the first persuasive message or

following the second persuasive message).

The results of the analysis were very supportive. The first

hypothesis predicted that individuals who expected highly intense

messages of a counterattitudinal nature, but whose expectations were

disconfirmed by the reception of low intensity messages, would

initially be more positive toward the attitude issue. It was further

predicted that upon receipt of a second message these individuals

would revert to being more negative. The second hypothesis predicted








that people initially induced to expect low intensity messages, but

whose expectations were disconfirmed by the reception of a highly

intense message, would be initially more negative toward the attitudinal

issue. Upon receipt of a second persuasive message, they were

predicted to become more positive toward the issue.

A significant three-way interaction with the pattern of the mean

attitude scores consistent with the predictions was required for

support of both of the hypotheses. The analysis indicated that the

hypothesized interaction was significant, and an inspection of the cell

means indicated that the obtained pattern matched the predicted rela-

tionships. Further support was provided for Hypothesis 1 by correlated

t-tests indicating that the experimental group expecting a highly

intense message but receiving a message of low intensity had signifi-

cantly more positive attitudes toward the issue after receipt of a

second persuasive message.

Taken together the results relevant to the theoretic hypotheses

provide strong support for the rationale advanced by this investigation.

It is evident that the support of Hypothesis 1 was crucial to this

study. While support of the second hypothesis alone would have

provided encouragement, the prediction that individuals expecting a

low intensity message and receiving a high intensity message would be

more positive toward the attitude issue after receipt of a second

message could have been made without reference to the rationale

advanced by this investigation. This effect could have been attributed

to a simple repetition effect.

As mentioned above, however, the finding that individuals

expecting highly intense messages and receiving a message of low








intensity become less favorable toward the issue after receipt of a

second persuasive message is much more exciting and much harder to

explain without reference to the extension of the Burgoon et al.

(1978) model presented here. This finding falls outside the domain

of both the inoculation analogy and congruity theory. Both of these

approaches deal strictly with pretreatment message strategies and make

no inferences about the possibility of one persuasive message affecting

resistance to subsequent persuasive messages advocating the same

attitudinal position. Moreover, the experimental and control group

comparisons negate the possibility of explaining this finding through

simple distraction or message repetition effects. All of the two-

message distraction control groups, as well as the two-message only

control groups, evidenced more positive attitudes after the receipt of

a second message. This finding, then, provides very strong support for

the rationale of the study.

The results of the analyses of the supplemental hypotheses are

less clear. The first supplemental hypothesis (Hypothesis 3) predicted

that perceived attitudinal threat would decrease after the presentation

of a second persuasive message. This hypothesis was tested with a 3

(receiver expectations of message intensity) by 2 (actual message

intensity) by 2 (time of measurement) repeated measures analysis of

variance. Although there was a trend toward significance of the

three-way interaction, no main effects or interactions obtained

significance. The failure to support a hypothesis mandates a

reexamination of both the procedures and methodology utilized in the

testing of the hypothesis and of the logic of the rationale leading

to the development of that hypothesis.








Initially, it was suggested that exposure to a highly counter-

attitudinal message would be very threatening. However, it was

reasoned that, given adequate time, a subsequent exposure to a second

persuasive message on the same attitudinal topic would prove less

threatening. This was suggested on the basis of two assumptions.

First, it was assumed that a second exposure to a counterattitudinal

message on the same topic would be more expected and therefore less

threatening to the individual. In addition, it was suggested that

time lapse between the initial exposure and the second exposure would

allow people the opportunity to rehearse counterarguments and more

adequately prepare their defenses.

The first assumption was tested by Hypothesis 5. This hypothesis

formalized the prediction that individuals would find the second

persuasive message more expected than the first. Since this hypothesis

received no support, an important antecedent to the consequent predicted

by Hypothesis 3 was not a factor in the design. This in itself reduces

the possibility of finding support for Hypothesis 3. In addition, the

assumption that the time lapse between exposures to the two persuasive

messages would allow for increased preparation and counterargumentation

may not have been appropriate in this investigation.

Although this assumption may be appropriate to many persuasive

situations, it may not have been extended far enough in the present

investigation. The rationale leading to the theoretic hypotheses

argued that negative violations of receiver expectations would lead to

a decrease in threat to the individual, while positive violations of

expectations would lead to increased threat to the individual. An

extension of this reasoning would suggest that an interaction effect








would have been a more appropriate prediction for this hypothesis.

While the analysis did indicate a trend toward a three-way interaction,

the interaction did not reach significance. There is also a problem

in specifying the exact nature of the expected interaction. This

problem arises in the lack of knowledge concerning the effects of a

second message on reduction of threat. This problem is compounded by

the consideration of the positive and negative violations of expecta-

tions which occurred by the presentation of the first persuasive

message. Considering both the methodological problems and the failure

to extend the rationale, the lack of support for this hypothesis is not

unexpected. However, the trend toward a three-way interaction does

provide some encouragement.

The rationale leading to Hypothesis 6 is very related to the third

hypothesis. Hypothesis 6 predicted a negative relationship between

threat induced by the first persuasive message and attitude scores

following a second persuasive attack. A significant negative correla-

tion between threat at Time 1 and attitudes at Time 2 provided support

for this hypothesis. This finding was important as a check on an

assumption of the rationale leading to the theoretic hypotheses. It

was assumed throughout this investigation that threat is an important

mediational variable in the resistance to persuasion process. Threat

is assumed to operate as a "trigger" to the counterargumentation

process. Thus the greater the threat induced by a first persuasive

message, the more likely the individual will counterargue in the

future, and the more likely that individual will resist a subsequent

persuasive attack. The support of this hypothesis provides empirical

substantiation for this assumption.







Hypothesis 4 predicted that individuals would express more

confidence in their attitudes after the receipt of a second persuasive

message. All subject groups receiving expectancy instructions and two

persuasive messages were utilized in the analysis of this hypothesis.

The three-way analysis of variance indicated a significant main effect

for time of measurement such that confidence in the attitude was

greater after receipt of a second persuasive message. No other main

effect clearly provides support for the hypothesis. This is consistent

with the reasoning, suggested by Burgoon et al. (1978), that after

deliberating on two messages and being allowed to provide defenses and

change or not change, people should be more confident in their

attitudes.

The results of the tests of the supplemental research questions

were varied. The effect of confirmations and disconfirmations of

expected message intensity on receiver perceptions of a source's

competence, composure, character, sociability and extroversion, and

on perceptions of a source of a subsequent message were tested. All

groups who received induced expectancy sets and both persuasive

messages were included in the analysis. Five separate three-way

analysis of variance tests were used to investigate the effects of the

independent variables on the different dimensions of credibility.

The only effects to obtain significance were time of measurement

main effects on the dependent variables of competence, composure and

character. The effects were such that the source of the first attack

message was seen as less competent, less composed and of lower

character than the source of the second persuasive attack. Since,

however, the same pattern of means appeared in the distraction control








conditions as in the experimental groups, the effects cannot be

unequivocally attributed to confirmations and disconfirmations of

receiver expectations. It may rather be a simple distraction effect

or, possibly, a contrast effect attributable to receiving a moderate

message subsequent to the reception of a high or low intensity counter-

attitudinal message.

The comparison of the experimental and control groups provided

manipulation checks and further support for the theoretic hypotheses.

A Duncan's multiple range test indicated that at Time 1, only the

experimental groups receiving a low intense message that either con-

firmed or disconfirmed their expectations differed significantly from

the pretest-posttest only control group. No predictions concerning

the reception of a low intense message which confirmed receiver

expectations were made, but the finding that people expecting a highly

intense message and receiving a low intense message would be initially

more positive is consistent with Hypothesis 1.

At Time 2, the Duncan's multiple range procedure indicated that

this group had reverted to a more negative attitudinal position and

was no longer significantly different from the pretest-posttest only

control group. This finding is also consistent with the first

hypothesis. Similarly, the group which had expected low intensity

language, but received high intensity language, was significantly

different from the pretest-posttest only control group at Time 2, even

though this difference was not evident at Time 1. This is consistent

with Hypothesis 2. The persuasive effectiveness of the combination

of two persuasive messages was evidenced by the significant differences

between each of the two-message only control groups and the








pretest-posttest only control group. The experimental condition

expecting and receiving low intense language remained significantly

different from the pretest-posttest only control condition at Time 2.

All of the one-message only control groups remained nonsignificantly

different from the pretest-posttest only control group at Time 2.

These experimental and control group comparisons are generally in the

pattern predicted by the rationale of this investigation, providing

additional support for Hypothesis 1 and Hypothesis 2.

Implications for Future Research and Conclusions

Burgoon, Cohen, Montgomery and Miller (1978) suggested that the

import of their new model of resistance to persuasion is the recon-

ceptualization of the resistance to persuasion process. They indicated

that their initial investigation provided evidence suggesting that

(1) any message might affect the persuasive efficacy of a subsequent

persuasive attack, and (2) the induction of a critical response set

may either inhibit or enhance the effects of a forthcoming message.

The primary import of the present investigation is the further support

and extension of this reconceptualization.

The present investigation extended the finding that critical

response sets may either inhibit or enhance the effectiveness of a

subsequent persuasive attack. The investigation provides evidence that

not only are the response sets themselves important, the systematic

violation of these receiver expectations may add to the enhancement or

inhibition of a second persuasive attack. The findings of the present

investigation also further support the assumption that any message

may affect the persuasive efficacy of a forthcoming persuasive attack.








On a broader level the investigation mandates an end to the

singular research emphasis on the pretreatment message strategies

suggested by the inoculation approach to the induction of resistance

to persuasion. While beginning the process, this investigation high-

lights the need and the utility of integrating findings from traditional

persuasion research into investigations of resistance to persuasion.

There are several areas in which this integration would appear to be

especially beneficial.

First, the message variables and other factors of the communication

situation which affect receiver perceptions of attitudinal threat need

to be more carefully investigated with the model advanced in this

investigation and the work of Burgoon et al. (1978). The rationale of

this line of research suggests that threat is an important variable in

the resistance to persuasion process because it serves as a "trigger"

for counterargumentation. This knowledge of factors which induce

attitudinal threat would allow the scope of this model to be broadened.

For example, Burgoon et al. (1974) have argued that intense language

usage by a female source constitutes a negative violation of expecta-

tions. It would seem, then, that the previous research of Burgoon et

al. (1974) on the persuasion process might have immediate application

in this new approach to resistance to persuasion.

Similarly, other types of violations of receiver expectations need

to be researched and applied to this model. The present investigation

considered violations of expected message intensity. It seems obvious,

however, that the disconfirmation of other types of receiver expecta-

tions might also enhance or inhibit subsequent message acceptance.








Furthermore, it seems evident that while the knowledge of the

persuasion process has application to this model, it can also be argued

that this model of resistance to persuasion may also enhance under-

standing of the traditional persuasion process. This seems most

evident concerning studies which utilize some type of message repeti-

tion. For example, Miller and Burgoon (1974) have pointed out that

the relationship between message repetition and attitude change

quickly become asymptotic. The concepts of attitudinal threat,

receiver expectations and counterargumentation might be used to provide

an explanation for these findings. Another area that might be rein-

vestigated in respect to these concepts concerns the relative effective-

ness of message primacy and recency. More consistency within the

research might be found using the concepts advanced in the present

investigation.

In summation, as Burgoon et al. (1978) point out, this model

avoids problems associated with relying solely on pretreatment message

structures as predictors of resistance to persuasion. Moreover, this

approach is more isomorphic with sequential message reception condi-

tions in which persuasion normally occurs. This will allow the

systematic integration of findings from traditional persuasion studies

with this model of resistance to persuasion. In addition, the concepts

used in this approach to resistance also have application to research

on the persuasion process.


























APPENDIX A
COVER STORY AND EXPECTANCY INDUCTIONS














Expectation of Low Intensity

The Department of Communication and Broadcasting is currently

engaged in a project to compare video-taped, audio-taped and written

messages on their effectiveness as means of information transfer.

We are asking that your class aid us in this research by evaluating

different aspects of written messages. Although other members of your

class may be doing different tasks, we would like you to read the

attached message and underline the statements that you consider to be

the least intense, that is, the most weakly worded statements in the

message.

PLEASE READ THE FOLLOWING VERY CAREFULLY

It is very important that everyone follow similar procedures in

the completion of these tasks. Therefore, please follow these instruc-

tions very carefully. After you finish reading these instructions,you

are to turn to the next page and begin reading the message. As you

read, you are to underline statements that you feel are weakly worded.

Underline as you read. After you have finished reading the message,

turn this booklet face down and await further instructions. Thank you

very much for your cooperation.








Expectation of High Intensity

The Department of Communication and Broadcasting is currently

engaged in a project to compare video-taped, audio-taped and written

messages on their effectiveness as means of information transfer.

We are asking that your class aid us in this research by evaluating

different aspects of written messages. Although other members of your

class may be doing different tasks, we would like you to read the

attached message and underline the statements that you consider to be

the most intense, that is, the most strongly worded statements in the

message.

PLEASE READ THE FOLLOWING VERY CAREFULLY

It is very important that everyone follow similar procedures in

the completion of these tasks. Therefore, please follow these instruc-

tions very carefully. After you finish reading these instructions, you

are to turn to the next page and begin reading the message. As you

read, you are to underline statements that you feel are strongly worded.

Underline as you read. After you have finished reading the message,

turn this booklet face down and await further instructions. Thank you

very much for your cooperation.








Expectation of Both High and Low Intensity

The Department of Communication and Broadcasting is currently

engaged in a project to compare video-taped, audio-taped and written

messages on their effectiveness as means of information transfer.

We are asking that your class aid us in this research by evaluating

different aspects of written messages. Although other members of your

class may be doing different tasks, we would like you to read the

attached message and underline the statements that you consider to be

the least intense, that is, the most weakly worded statements in the

message, and to circle the statements that you consider to be the most

intense, that is, the most strongly worded statements in the message.

PLEASE READ THE FOLLOWING VERY CAREFULLY

It is very important that everyone follow similar procedures in

the completion of these tasks. Therefore, please follow these instruc-

tions very carefully. After you finish reading these instructions, you

are to turn to the next page and begin reading the message. As you

read, you are to underline statements that you feel are weakly worded

and circle statements that you feel are strongly worded. Underline and

circle as you read. After you have finished reading the message, turn

this booklet face down and await further instructions. Thank you very

much for your cooperation.




























APPENDIX B
EXPERIMENTAL MESSAGES














Low Intensity: Health Related


One concern of our society is the health and welfare of its
citizens. This concern has resulted in legislation which has provided
acceptable health care programs for the elderly, the poor and the
unemployed. While these programs have benefited many people, several
segments of our society remain in need of adequate health care. In
fact, one of the groups which needs care as much as many others, heroin
users, has been helped only slightly by current policies. Occasionally,
the laws regulating the sale of heroin in this country have done
almost as much harm as good. While some arguments concerning the
legalization of heroin sales involve issues other than health care,
we feel that our major focus should be on health-related issues.

Present criminal sanctions against the sale of heroin were, in
part, designed to protect the American citizen from the medical harms
which were thought to be caused by the drug. A possible result of some
of those laws, however, is that a number of heroin addicts are in poor
health from diseases caused by the drug's continued illegality. For
example, in 1969, over 900 addicts in New York City were treated for
tetanus and hepatitis. These diseases were not caused by heroin, but
were the result of improper means of injection. Since hypodermic
syringes cannot be obtained legally, some users choose to re-use and
share needles. Thus the laws which ban the sale of heroin probably
cause a number of heroin-related diseases each year.

A second health problem faced by the user is the purchase of
impure heroin. Because the drug is available primarily on the illicit
market, it is often prepared by street dealers who may or may not have
much concern for the health of their clients. Since users are occasion-
ally unsure of the amount or quality of the heroin they purchase, they
are sometimes unable to regulate the exact dosage of the drugs they
take. Perhaps as a consequence, a small percentage of heroin users may
accidentally harm themselves from overdosages. Some people have
speculated that if heroin users know the actual strength of the drug
they are using, medical problems caused by drug overdose might be
reduced. If the sale of heroin were legalized, the government might
be in a position to try and enforce some type of quality control on the
heroin sold, perhaps reducing the number of heroin-related diseases
each year.

Since the cost of illegal heroin is rather high, users are some-
times unable to afford supplemental items for use in maintaining good
health. Addicts' failure to get supplemental nutrition, medication and
doctors' care has been linked to their choosing to use much of their
resources to pay inflated black market prices for the drug. Under a








system of legalized sales the price of the drug might be significantly
reduced, and users would be better able to afford other health-related
products.

Another issue that must be considered is whether the use of heroin
constitutes a severe health problem. Some medical authorities now
agree that heroin use causes little physical damage. In addition,
researchers have shown that the symptoms of heroin withdrawal are no
more dangerous than those assocaited with the withdrawal from alcohol.
Therefore, when assured of the legal supply of the drug, the heroin
user is perfectly capable of leading a meaningful and productive life.

Various arguments can be used to support the legalized sale of
heroin. We feel, however, that even considering only the health care
benefits, eliminating secondary infection, reducing the number of over-
doses, and increasing the availability of nutrition and medication,
coupled with the evidence that heroin itself causes little physical
damage, justify the legalization of the sale of heroin in the United
States.








Low Intensity: Crime Related

Most responsible citizens realize that something must be done to
curb the rising crime rate in this country. What most people do not
realize, however, is that in our major cities a small percentage of the
crime reported is committed by heroin users seeking money to support
their habit.

A prominent journal recently noted that in England, where the
government controls the legal sale to heroin addicts, heroin-related
crimes have been somewhat reduced. In short, the relationship between
heroin use and crime may not be entirely caused by the drug itself, but
rather may be at least partially the result of some of the laws which
prohibit its use. Those who support the legalization of heroin argue
from several acceptable points of view. However, we feel that the
arguments related to crime are of average importance.

While we do not encourage the use of heroin, we still must view
the problem realistically. In this society we tolerate the use of
alcohol, tobacco and other drugs in the acceptable belief that individ-
uals should have some freedom in choosing their own life styles.
Strictly speaking, heroin use itself only harms the user. However, the
fact that it is illegal could possibly result in its harming not only
the user, but a few others as well.

Since heroin cannot be obtained legally, addicts sometimes choose
to commit petty theft and other minor crimes against people to get
money to pay high black market prices for the drug. Moreover, they
occasionally turn to individuals who may or may not be connected with
the underworld for their supply of heroin, possibly creating a profit
for "organized crime."

Thus, legalizing heroin sales might possibly reduce crime in two
ways. First, if heroin were legalized, it would be less expensive and
the addict might not rob innocent people for money to pay high prices
for the drug. Secondly, it would help in the fight against organized
crime by reducing the profit on one source of the underworld's income.

Another often neglected fact is that the illegal sale of heroin is
one of many sources of police corruption. Several independent com-
missions have pointed out that the trafficking of heroin is becoming a
fairly common and somewhat lucrative source of graft available to
policemen. Consequently, current laws are counterproductive; they are
somewhat poor at reaching the goal of reducing drug-related crime and
might possibly increase the amount of corruption in our police depart-
ments by turning them into illegal suppliers of heroin.

A major goal of any program concerning heroin use should be the
rehabilitation of the heroin addict. However, several researchers have
concluded that the fear of arrest may keep a few addicts from seeking
help. Thus the laws making heroin illegal are a factor in preventing
these few addicts from undergoing rehabilitation treatment. Putting




60


the addict in prison is not always the answer, either, since heroin is
occasionally almost as easy to obtain in a few prisons as it is in some
cities. It is clear that our current policies do not contribute as
much as they could to the rehabilitation of the heroin addict.

Obviously we as a nation cannot condone the use of heroin, and
certainly there are other OK points which support its legalization.
However, we feel that reducing the crime rate, fighting organized
crime, reducing one source of possible police corruption, and aiding
in the rehabilitation of heroin addicts are fair reasons that provide
some justification for a policy of legalizing heroin sales in the
United States.








Moderate Intensity: Health Related

One of the major concerns of our society is the health and welfare
of its citizens. This concern has resulted in legislation which has
provided specialized health care programs for the elderly, the poor and
the unemployed. While these programs have benefited many people,
several segments of our society remain in desperate need of adequate
health care. In fact, one of the groups which needs care the most,
heroin users, has actually been helped the least by current policies.
More often than not, the laws regulating the sale of heroin in this
country have done more harm than good. While many arguments concerning
legalization of heroin sales involve issues other than health care, we
feel that our major focus should be on health-related issues.

Present criminal sanctions against the sale of heroin were, in
part, designed to protect the American citizens from the medical harms
which were once thought caused by the drug. The result of those laws,
however, is that many heroin addicts are dying needlessly from diseases
caused not by heroin, but from secondary complications which are
promoted by the drug's continued illegality. For example, in 1969,
over 900 addicts died in New York City from tetanus and hepatitus.
These deaths were not caused by heroin, but were the result of improper
means of injection. Since hypodermic syringes cannot be obtained
legally, many users are forced to re-use and share needles, or they
improvise with objects not designed for injecting drugs into the
bloodstream. Thus the laws that ban the sale of heroin actually cause
a significant number of heroin-related deaths each year.

A second health hazard faced by the user is the purchase of
impure heroin. Because the drug is available only on the illicit
market, it is haphazardly prepared by street dealers who have little
concern about the health of their clients. Since users are never sure
of the amount or quality of the heroin they purchase, they are often
unable to regulate the dosage of the drugs they take. Consequently,
many heroin users accidentally die each year from drug overdose.
Research has indicated that when heroin users know the actual strength
of the drug they're using, deaths and medical problems caused by over-
dosing are virtually eliminated. If the sale of heroin were legalized,
the government would be in a position to enforce quality controls on
the heroin sold, thus saving many lives each year.

Since the cost of illegal heroin is so excessive, users are often
unable to afford items essential for maintaining good health. Addicts'
failure to get needed nutrition, medication and doctors' care has been
directly linked to their being forced to use all of their resources to
pay inflated black market prices for the drug. Under a system of
legalized heroin sales, the price of the drug would be greatly reduced,
and users could afford other essential health-related products.

Another issue that must be considered is whether the use of heroin
constitutes a real health problem. Medical authorities now agree that
heroin causes little physical damage. In addition, researchers have




62


shown that the symptoms of heroin withdrawal are much less dangerous
than those associated with the withdrawal from alcohol. Therefore,
when assured of the legal supply of the drug, the heroin user is
capable of leading a meaningful and productive life.

Various arguments can be used to support the legalized sale of
heroin. We feel, however, that even considering only the health care
benefits, eliminating secondary infection, reducing the number of
overdoses, and increasing the availability of nutrition and medication,
coupled with the evidence that heroin itself causes little physical
damage, justify the legalization of the sale of heroin in the United
States.








Moderate Intensity: Crime Related

Most responsible citizens realize that something must be done to
curb the rising crime rate in this country. What most people do not
realize, however, is that in our major cities a high percentage of the
crime reported is committed by heroin addicts seeking money to support
their habit.

A prominent journal recently noted that in England, where the
government controls the legal sale to heroin addicts, heroin-related
crimes are almost nonexistent. In short, the relationship between
heroin use and crime is not caused by the drug itself, but rather by
the laws which prohibit its use. Those who support the legalization
of heroin argue from several points of view. However, we feel that the
arguments related to crime are of primary importance.

While we do not encourage the use of heroin, we still must view
the problem realistically. In this society we tolerate the use of
alcohol, tobacco, and other drugs in the belief that individuals should
have considerable freedom in choosing their own life styles. Strictly
speaking, heroin use itself only harms the user. However, the fact
that it is illegal results not only in its harming the addict, but
many others as well.

Since heroin cannot be obtained legally, addicts are forced to
commit crimes against many innocent people to get money to pay high
black market prices for the drug. Moreover, they must turn to the
underworld for their supplies of heroin, making organized crime richer
and more powerful. Thus, legalizing heroin sales would reduce crime
in at least two ways. First, if heroin were legalized, it would be
relatively inexpensive, and the addict would not be forced to prey upon
innocent citizens for money to pay exorbitant prices for the drug.
Secondly, it would help in the fight against organized crime by taking
away an important source of the underworld's income.

Another often neglected fact is that the illegal sale of heroin is
a prime source of police corruption. Several independent commissions
have pointed out that the trafficking of heroin is one of the most
common lucrative sources of graft that is available to policemen.
Consequently, current laws are counterproductive; they undermine the
goal of reducing drug-related crime and increase the possibility of
corruption in our police departments by turning them into illegal
suppliers of heroin.

A major goal of any program concerning heroin use should be the
rehabilitation of the heroin addict. However, several researchers
have concluded that the fear of arrest keeps many addicts from seeking
help. Thus, the laws making heroin illegal are actually preventing
rehabilitation of many of those addicted to the drug. Putting the
addict in prison is no answer, either, since heroin is often easier to
obtain in prison than on the streets. It is clear that our current
policies contribute little to the rehabilitation of the heroin addict.




64


Obviously, we as a nation cannot condone the use of heroin, and
certainly there are other points which support its legalization.
However, we feel that reducing the crime rate, fighting organized
crime, reducing a major source of police corruption, and aiding in the
rehabilitation of heroin addicts are reasons that provide sufficient
justification for a policy of legalizing heroin sales in the United
States.








High Intensity: Health Related

One of the major concerns of our society is the health and welfare
of its citizens. This concern has resulted in legislation which has
provided specialized health care programs for the elderly, the poor and
the unemployed. While these programs have benefited many people,
several segments of our society remain in critical need of adequate
health care. In fact, the one group which probably needs care the most,
heroin users, has actually suffered irreparable damage under the
terrible current policies. In every instance, the laws regulating the
sale of heroin in this country have done more harm than good. While
many excellent arguments concerning legalization of heroin sales
involve issues other than health care, we feel that our major focus
should be on health-related issues.

Terrible criminal sanctions against the sale of heroin were, in
part, designed to protect the American citizen from the despised
medical harms which were once thought caused by the drug. The result
of those very bad laws, however, is that thousands of heroin addicts
are dying needlessly from horrible diseases caused not from heroin,
but from deadly secondary complications which are the direct result of
the drug's continued illegality. For example, in 1969, over 900
addicts suffered agonizingly slow deaths from tetanus and hepatitus.
These deaths were not caused by heroin, but were the result of improper,
almost medieval, means of injection. Since hypodermic syringes cannot
be obtained legally, many users are literally forced to re-use and
share needles--or they improvise with filthy objects not designed for
injecting drugs into the bloodstream. Thus the laws which ban the sale
of heroin actually cause an enormous number of heroin-related deaths
each year.

A second health hazard faced by the user is the purchase of
terribly impure heroin. Because the drug is available only on the
illicit black market, it is haphazardly prepared by pushers who have
absolutely no concern about the health of their clients. Since users
are never sure of the amount of quality of the heroin they purchase,
they are never able to regulate the dosage of the drugs they take.
Consequently many heroin users accidentally die each year from drug
overdose. Research has indicated that when heroin users know the
actual strength of the drug they are using, the horrible deaths caused
by overdosing are virtually eliminated. If the sale of heroin were
legalized, the government would be in an excellent position to enforce
quality controls on the heroin sold, thus saving many lives each year.

Since the cost of illegal heroin is so excessive, addicts are
almost never able to afford the bare necessities essential for main-
taining even adequate health. Addicts' failure to get needed nutrition,
medication, and doctors' care has been a direct result of their being
forced to use all of their resources to pay inflated black market
prices for the drug. Under a system of legalized heroin sales, the
price of the drug would be greatly reduced, and addicts could afford
other essential health-related products.




66


Another issue that must be considered is whether the use of
heroin constitutes a real health problem. Medical authorities now
agree that heroin use causes no physical damage. In addition,
researchers have shown that the symptoms of heroin withdrawal are much
less dangerous than those associated with the withdrawal from alcohol.
Therefore, when assured of the legal supply of the drug, the heroin
user is perfectly capable of leading a meaningful and productive life.

Various excellent arguments can be used to support the legalized
sale of heroin. We feel, however, that even considering only the
health care benefits, eliminating secondary infection, reducing the
number of overdoses, and increasing the availability of nutrition and
medication, coupled with the evidence that heroin itself causes no
physical damage, justify the legalization of the sale of heroin in the
United States.








High Intensity: Crime Related

Every responsible citizen realizes that something must be done to
curb the drastically rising crime rate in this country. What most
people do not realize, however, is that in our major cities an
extremely high proportion of the crime reported is committed by heroin
addicts seeking money to support their habit.

A prominent journal recently noted that in England, where the
government controls the legal sale of heroin to addicts, heroin-
related crimes have become nonexistent. In short, the relationship
between heroin use and crime is not a function of the drug, but is
rather caused by the archaic laws which prohibit its use. Those who
support the legalization of heroin argue from several excellent points
of view. However, we feel that the arguments related to crime are the
best of all and are of primary importance.

While we do not encourage the use of heroin, we still must face
the cold, hard facts. In this society we tolerate the use of alcohol,
tobacco, and other drugs in the knowledge that individuals must have
considerable freedom in choosing their own life styles. It is a well-
known fact that heroin laws that prevent its legal sale result not only
in harming, perhaps even killing, the addict but many others as well.

Since heroin cannot be obtained legally, addicts are forced to
commit heinous crimes against completely innocent people to get money
to pay high black market prices for the drug. Moreover, they must turn
to despicable underworld figures for their supplies of heroin--thus
making organized crime richer and even more powerful. Thus, legalizing
heroin sales would significantly reduce crime in at least two ways.
First, if heroin were legalized, it would be much less expensive and
the addict would not be forced to relentlessly prey upon the innocent
citizenry of our country for money to pay exorbitant prices for the
drug. Secondly, it would enormously help in the fight against
organized crime by taking away an important, irreplaceable source of
the underworld's income.

Another often neglected fact is that the illegal sale of heroin is
a favorite, prime source of police corruption. Several independent
commissions have discovered that the trafficking of heroin is the
single most common and most lucrative source of graft available to
policemen. Consequently, current laws are counterproductive; they
undermine the goal of reducing drug-related crime and drastically
increase the possibility of corruption in our police departments by
turning policemen into illegal dealers of heroin.

A major goal of any program concerning heroin use should be the
rehabilitation of the heroin addict. However, several researchers
have concluded that the fear of a dreaded arrest and severe prison term
keeps many addicts from seeking help. Thus the laws making heroin
illegal are actually preventing rehabilitation of many of those
addicted to the crime. Putting the addict in prison is no answer,
either, since heroin is actually much easier to obtain in prison than




68


on the streets. It is clear that rather than contributing to the
rehabilitation of the heroin addict, our current policies help condemn
them to the pitiable life of a junkie.

Obviously, we as a nation cannot condone the use of heroin, and
certainly there are other excellent points which support its legaliza-
tion. However, we feel that reducing the crime rate, fighting
organized crime, reducing a major source of police corruption, and
aiding in the rehabilitation of heroin addicts are reasons that provide
more than sufficient justification for a policy of legalizing heroin
sales in the United States.




























APPENDIX C
MEASUREMENT INSTRUMENT













PLEASE READ THE FOLLOWING VERY CAREFULLY


On the following pages you will find a series of statements and objects
followed by several scales. Please mark each scale in the blank that
BEST represents how you feel about the statement or object. For
example, here is an item like the ones you will see:

The United States should withdraw from the United Nations.

Good : : : : : : :Bad

Your job is to place a check mark (X) above the line that best
indicates your feeling toward the statement. For example, if you feel
that U.S. withdrawal would be a very good idea, you would check as
follows:

Good X : : : : : : :Bad

If you feel such a move (withdrawal) would be slightly beneficial, you
would check as follows:

Good : : X : : : : :Bad

If you feel neutral or indifferent about the proposition, or if you
feel that particular scale is irrelevant to the proposition, you would
check as follows:

Good : : : X : :BBad

Remember: Fill out every scale and never make more than one mark on a
single scale. Thank you for your cooperation.




71


1. The sale of heroin should be legalized in the United States.

Bad : : : :_ :Good
Valuable _ __ :Worthless
Foolish__ : : :_1Wise
Pleasant___ __ __ Unpleasant

2. Please indicate your reactions to the communication you have just
read.

Safe__ : : __ :Dangerous
Predictable __ :Unpredictable
Threatening_ : Not threatening
Unexpected_ : Expected
Risky__ : . ._ :Not risky
Surprising__ __ __ __ : lot surprising
Intimidating_ : Unintimidating
Unlikely :Likely
Harmful l_ __ ot harmful
Anticipated_ :Unanticipated

3. Please indicate your reactions to the author of the message you
just read.

Cheerful :Gloomy
Cruel :__ Kind
Tense_ : : : : : :Relaxed
Verbal ______________Quiet
Reliable :Unreliable
Sociable_ :Unsociable
Nervous_ : : : :_ Poised
Selfish_ :Unselfish
Meek__ __ _: Aggressive
Expert_ __ : Inexpert
Timid__ :Bold
Good-natured_ : Irritable
Calm_ :Anxious
Sympathetic_ :Unsympathetic
Unqualified :Qualified

4. Please indicate how you feel about the decisions you have made
during this survey.

Right : : :__:____ :Wrong
Confident_ :Not confident
Certain____ : : :Uncertain
Negative_ :Positive
Sure_ :__ Unsure













REFERENCES


Baron, R.S., Baron, P., & Miller, N. The relation between distraction
and persuasion. Psychological Bulletin, 1973, 80, 310-323.

Becker, S.W., Bavelas, A., & Braden, M. An index to measure contin-
gency of English sentences. Language and Speech, 1961, 4,
139-145.

Bowers, J.W. Language intensity, social introversion, and attitude
change. Speech Monographs, 1963, 30, 345-352.

Bowers, J.W. Some correlates of language intensity. Quarterly
Journal of Speech, 1964, 50, 415-420.

Breitrose, H.S. The effect of distraction in attenuating counter-
arguments. Doctoral dissertation, Stanford University, 1966.

Brooks, R.D. The generality of early reversal of attitudes toward
communication sources. Speech Monographs, 1970, 37, 152-155.

Brooks, R.D., & Scheidel, T.M. Speech as a process: A case study.
Speech Monographs, 1968, 37, 1-7.

Burgoon, M. The effects of response set and race on message inter-
pretation. Speech Monographs, 1970, 37, 264-268.

Burgoon, M., Burgoon, J., Reiss, M., Butler, J., Montgomery, C.,
Stinnett, W., Miller, M., Long, M., Vaughn, D., & Caine, B.
Propensity of persuasive attack and intensity of pretreatment
messages as predictors of resistance to persuasion. Journal of
Psychology, 1976, 92, 123-129.

Burgoon, M., & Chase, L.J. The effects of differential linguistic
patterns in messages attempting to induce resistance to persua-
sion. Speech Monographs, 1973, 40, 1-7.

Burgoon, M., Cohen, M., Montgomery, C.L., & Miller, M.D. An empirical
test of a model of resistance to persuasion. Human Communication
Research, 1978, in press.

Burgoon, M., Jones, S.B., & Stewart, D. Toward a message-centered
theory of persuasion: Three empirical investigations of language
intensity. Human Communication Researcn, 1974, 1, 240-256.








Burgoon, M., & King, L.B. The mediation of resistance to persuasion
strategies by language variables and active-passive participation.
Human Communication Research, 1974, 1, 30-41.

Burgoon, r., & Miller, G.R. Prior attitude and language intensity as
predictors of message style and attitude change following counter-
attitudinal advocacy. Journal of Personality and Social
Psychology, 1971, 20, 240-253.

Clark, H.H., & Begun, J.S. The semantics of sentence subjects.
Language and Speech, 1971, 14, 34-36.

Festinger, L., & Maccoby, N. On resistance to persuasive communications.
Journal of Abnormal and Social Psychology, 1964, 68, 359-366.

Freedman, J.L., & Sears, D.O. Warning, distraction, and resistance to
influence. Journal of Personality and Social Psychology, 1965,
1, 262-266.

Gardner, D.M. The effect of divided attention on attitude change
induced by a persuasive marketing communication. In Raymond H1.
Haus (Ed.), Science, technology, and marketing. Chicago,
American Marketing Association, 1966.

Haaland, G.A., & Venkatesan, M. Resistance to persuasive communica-
tions: An examination of the distraction hypothesis. Journal of
Personality and Social Psychology, 1968, 9, 167-170.

Insko, C.S., Turnbull, Wl., & Yandell, B. Facilitative and inhibitive
effects of distraction on attitude change. Sociometry, 1974, 37,
508-528.

Kiesler, S.B., & Hlathog, R.B. Distraction hypothesis in attitude
change: Effects of effectiveness. Psychological Reports, 1968,
23, 1123-1133.

rcCroskey, J.C., Jensen, T., & Todd, C. The generalizability of
source credibility scales for public figures. Paper presented at
the Speech Communication Association Convention, Chicago, 1972.

McEwen, W.J. Counterattitudinal encoding effects on message style and
performance. Doctoral dissertation, Michigan State University,
1969.

MicGuire, W.J. Resistance to persuasion conferred by active and passive
refutation of the same and alternative arguments. Journal of
Abnormal and Social Psychology, 1961, 63, 326-332.

McGuire, W.J. Threat and reassurance as factors in conferring
resistance to persuasion. Unpublished manuscript, Columbia
University, 1963. Described in W.J. McGuire, Inducing resistance
to persuasion: Some contemporary approaches. In L. Berkowitz
(Ed.), Advances in experimental social psychology, Vol. 1. New
York: Academic Press, 1964, 191-229.







McGuire, W.J. Inducing resistance to persuasion: Some contemporary
approaches. In L. Berkowitz (Ed.), Advances in experimental
social psychology, Vol. 1, New York: Academic Press, 1964,
191-229.

McGuire, W.J. Attitudes and opinions. Annual Review of Psychology,
1966, 17, 475-514.

McGuire, W.J. The nature of attitudes and attitude change. In
G. Lindzey & E. Aronson (Eds.), Handbook of social psychology,
Vol. III. Reading, Mass.: Addison-Wesley, 1969, 136-314.

McGuire, W.J., & Papageorgis, D. Effectiveness of forewarning in
developing resistance to persuasion. Public Opinion Quarterly,
1962, 26, 24-34.

Miller, G.R., & Burgoon, M. New techniques of persuasion. New York:
Harper and Row, 1973.

Miller, M.D., & Burgoon, M. Overcoming resistance to persuasion via
reinforcement and campaign techniques. Paper presented to the
Speech Communication Association Convention, Chicago, 1974.

Miller, M.D., & Burgoon, M. Predictors of resistance to persuasion:
Propensity of persuasive attack, pretreatment language intensity,
and expected delay of attack. Journal of Psychology, 1977, 95,
105-110.

Osterhouse, R.A., & Brock, T.C. Distraction increases yielding to
propaganda by inhibiting counterarguing. Journal of Personality
and Social Psychology, 1970, 15, 344-358.

Papageorgis, D., & McGuire, W.J. The generality of immunity to persua-
sion produced by pre-exposure to weakened counterarguments.
Journal of Abnormal and Social Psychology, 1961, 62, 475-481.

Regan, D.T., & Cheng, J.B. Distraction and attitude change: A
resolution. Journal of Experimental Social Psychology, 1973, 9,
138-147.

Rosenblatt, P.C. Persuasion as a function of varying amounts of
distraction. Psychonomic Science, 1966, 5, 85-86.

Tannenbaum, P.H. The congruity principle revisited: Studies in the
reduction, induction and generalization of attitude change. In
L. Berkowitz (Ed.), Advances in experimental social psychology,
Vol. III. New York: Academic Press, 1967, 271-320.

Vohs, J.L., & Garrett, R.L. Resistance to persuasion: An integrative
framework. Public Opinion Quarterly, 1968, 32, 445-452.




75


Zimbardo, P.G., Snyder, N., Thomas, J., Gold, A., & Gurwitz, S.
Modifying the impact of persuasive communications with external
distraction. Journal of Personality and Social Psychology, 1970,
16, 669-680.













BIOGRAPHICAL SKETCH


Michael D. Miller was born September 19, 1952, at the Army-Navy

Hospital in Hot Springs, Arkansas, to Henry T. and Billie Jean Miller.

Contrary to a recurring rumor, it is only coincidence that the hospital

of his birth was shortly thereafter converted to the Arkansas Rehabil-

itation Center. Most of his childhood was spent in such august

metropolises as Pocahontas, Arkansas, and Gurdon, Arkansas. The rumor

that during this period of Mr. Miller's life he was so ugly his mother

had to tie a pork chop around his neck so the dogs would play with him

is also untrue.

Mr. Miller's graduation from Gurdon High School in 1970 evidenced

the first point of a linear relationship between the level of degree

sought and the quality of the institution attended by Mr. Miller. No

comment is made on whether this relationship is direct or inverse.

In May 1973, after learning more than he ever wanted to know about

diagraming sentences and how to maintain order in public school lunch-

rooms, Mr. Miller received the degree of Bachelor of Science in

Education from Henderson State College in Arkadelphia, Arkansas.

Continuing his tradition of attending only renowned centers of

excellence in higher education, Mr. Miller began studies for the

Master of Arts degree in the Department of Speech-Communication at

West Virginia University. During his brief tenure with the Department

of Speech-Communication, Mr. Miller learned many things from several








distinguished faculty, not the least of which were appropriate

convention behavior and appropriate use of factor analysis. However,

all things, both good and bad, must come to an end. In August of 1974

Mr. Miller received the Master of Arts degree from West Virginia

University.

Not having satisfied his thirst for knowledge, Mr. Miller donned

his Mae West and began studies on the Flagship, the University of

Florida. The most significant aspects of his learning experiences on

board the Flagship concerned aspects of professionalism and ethics.

In February of 1977 Mr. Miller wed the former Debra Bennett.

It should be noted that at the time of the ceremony he was not wearing

a pork chop necklace. They have one son, Andrew. In the summer of

1977, with professionalism at its zenith on board the Flagship, the

Miller clan moved to Laramie, Wyoming, where Mr. Miller joined the

faculty of the Department of Communication at the University of

Wyoming.

Returning to the Flagship for the last time, Mr. Miller completed

the requirements for the degree of Doctor of Philosophy in the summer

of 1978.

At last report Michael, Debra and Andrew were going home to a

place they had seldom been before, Hawaii. There Dr. Miller will

combine the occupation of Assistant Professor of Communication at the

University of Hawaii with that of beachbum.












I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Michael Burgoon, i.D., Chairman*
Associate Professor of Communication
Michigan State University


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.


I /' .-
*-. . ,,. j ^, -, ,' -,'L ,' ^- .__
Judee K. Burgoorn/;Ed.D.**
Assistant Profesbor of Communication
Michigan State University


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Norman N. Markel, Ph.D.
Professor of Speech


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




G. Paul Moore, Ph.D.
Professor of Speech

*Formerly Associate Professor, Speech. University of Florida
**Formerly Assistant Professor, Speech, University of Florida












I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.


Professor of Psychology


This dissertation was submitted to the Graduate Faculty of the
Department of Speech in the College of Arts and Sciences and to
the Graduate Council, and was accepted as partial fulfillment of
the requirements for the degree of Doctor of Philosophy.

August 1978


Dean, Graduate School




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs